TWI272737B - Method and system for joint battery state and parameter estimation - Google Patents

Method and system for joint battery state and parameter estimation Download PDF

Info

Publication number
TWI272737B
TWI272737B TW093136725A TW93136725A TWI272737B TW I272737 B TWI272737 B TW I272737B TW 093136725 A TW093136725 A TW 093136725A TW 93136725 A TW93136725 A TW 93136725A TW I272737 B TWI272737 B TW I272737B
Authority
TW
Taiwan
Prior art keywords
prediction
state
internal
uncertainty
unit
Prior art date
Application number
TW093136725A
Other languages
Chinese (zh)
Inventor
Gregory L Plett
Original Assignee
Lg Chemical Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Lg Chemical Ltd filed Critical Lg Chemical Ltd
Priority claimed from PCT/KR2004/003102 external-priority patent/WO2006057469A1/en
Application granted granted Critical
Publication of TWI272737B publication Critical patent/TWI272737B/en

Links

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
    • H01M10/00—Secondary cells; Manufacture thereof
    • H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/374—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] with means for correcting the measurement for temperature or ageing
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
    • G01R31/3842—Arrangements for monitoring battery or accumulator variables, e.g. SoC combining voltage and current measurements
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
    • Y02E60/10—Energy storage using batteries

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Chemical & Material Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Electrochemistry (AREA)
  • General Chemical & Material Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Secondary Cells (AREA)

Abstract

A method and apparatus for estimation of the augmented state of an electrochemical cell, the method comprising: making an internal augmented states prediction of the cell where the augmented state comprises at least one internal state value and at least one internal parameter value; making an uncertainty prediction of the internal augmented states prediction; correcting the internal augmented states prediction and the uncertainty prediction; and applying an algorithm that iterates the making an internal augmented states prediction, the making an uncertainty prediction and the correcting to yield an ongoing estimation to the augmented state and an ongoing uncertainty to the augmented state estimation.

Description

1272737 九、發明說明： 【發明所屬之技術領域】 本發明係關於一種利用數位濾波技術來評估電池組 系統狀態及模組參數的方法與裝置，尤指一種利用聯合卡 5 爾曼遽波法（joint Kalman filtering)及聯合延伸卡爾曼濾、波 法（joint extended Kalman filtering)之電池組系統狀態及模 組參數的評估裝置及方法。 【先前技術】 10 按，於可充電電池組（rechargeable battery pack)之技 術領域中，較佳皆須具備可評估並描述電池組目前狀態的 應用功能。然而，實際上有些電池組的目前狀態值並無法 直接被量測出。況且，其中部分的狀態值改變得相當迅速， 例如電池組的充電狀態（state of charge，S0C)，其改變幅度 15 可於短暫幾分鐘内達到其完整的應用範圍；反之，部分的 狀態值則改變得相當地緩慢，例如電池電荷容量（cell capacity)在數十年間之正常使用的前提下，其改變幅度僅 約20%而已。一般來說，那些傾向於迅速改變的狀態值包 括系統的狀態（state)，而那些傾向於緩慢改變的狀態值則 20 包括系統中那些會隨著時間變化的參數（parameter)。 而於電池系統之技術領域中，諸如複合式電動車 (hybrid electric vehicles，HEVs)、電池電動車（battery electric vehicles，BEVs)、筆記型電腦電池、可攜式工具電 池組及其他等效且需儘可能強效地長時間運作而不會傷害 1272737 八電池可〒的電池系統中，較佳係利用那些與快速改變之. 多數（例如充電狀態參數）有關的資訊評估一電池目前尚餘 夕 > 可用電力。此外，較佳亦需確認那些與慢速改變之參 數(例如總電荷容量參數）有關的資訊，俾確保在電池組之 5壽命週期中，冑能得到精準的先前計算、延長電池組的服 務時效並協助判定電池組的健康狀態（state h^lth， SOH) 〇 ’ 現存許多可評估電池狀態的方法，其大致皆係評估下 列二種狀態值：充電狀態（SOC)、電力衰退（power fade)及 10電荷容量衰退（capacity fade)，其中充電狀態係為一快速改 變的狀態值，而電力衰退及電荷容量衰退則皆屬於慢速改 k之狀態值。此外若知道目前及初始電池組之電阻，則可 计算出電力衰退；若已知目前及初始電池組之總電荷容 里，則可計算出電荷容量衰退。當然，電力衰退及電荷容 15量衰退亦可利用其他不同的方法計算出，並不限於上述之 方法。當描述一電池組之健康狀態（s〇H)時，電力衰退與 電％谷i衣退常被混為一談。利用上述之變數值可推導出 些其他資訊’如電池組在任一時間點的最大可用電力 (maximum available power)。此外，在某些特定應用中， 20舄使用到額外的狀態構件（state member)或參數，而這些狀 態構件或參數需使用個別的計算法求出。 充電狀態（SOC)係一數值，其通常以百分比方式表 不，俾指出電池組目前尚餘之可運用的電力。目前有許多 不同的方法被用於評估一電池組之充電狀態，包括：放電 1272737 測試法（discharge test)、安培小時計數法（ampere-hour counting)(或稱庫命計算法（Coulomb counting))、量測電解 液法（measuring electrolyte)、開路電壓量測法（open-circuit voltage measurement)、線性及非線性電路模型化法（linear 5 and nonlinear circuit modeling)、阻抗頻譜分析法 (impedance spectroscopy)、内部電阻量測法、電壓驟降法 (Coup de fouet)及某些形式的卡爾曼濾波器等。其中，放 電測試法必須將電池完全放電以判斷充電狀態，且於執行 放電測試時必需中斷系統功能並耗時甚久。因此，放電測 10 試的應用範圍有限。安培-小時計數法（或庫侖計算法）係採 用「開放式迴圈」的方法，其精確度會因量測誤差值的累 積而隨著時間而減低。量測電解液法僅適用於開口型鉛酸 電池（vented lead-acid battery)，其應用性相當有限。開路 電壓量測法僅適用於具有可忽略磁滯（hysteresis)效應的電 15 池上，且僅在電池處於長時期未使用的狀態下才可應用。 此外，開路電壓量測法並無法在動態環境下運作。線性及 非線性電路模型化方法並無法直接地計算出電池之充電狀 態，電池之充電狀態必須由計算出的數值加以推算而得 知。阻抗頻譜分析法所需進行的量測並非一般應用可達 20 成。内部電阻量測法對於量測誤差值非常地敏感’且所需 進行的量測在一般的應用中並無法達成。電壓驟降法僅適 用於鉛酸電池中。那些不將充電狀態（SOC)作為濾波器狀 態的某些卡爾曼濾波法形式並無法直接計算出量測值的誤 差範圍（error bound)。而於其他方法中’例如美國專利公 1272737 告第6,534,954號（在本文中以參考文獻方式全面地含括了 该專利），係藉由運用一已知之電池動力學的數學模型以及 電池之電壓、電流及溫度的量測值，透過一濾波器（較佳為 卡爾曼;慮波器）#估電池之充電狀態。然而，雖然此方法 5可直接評估狀態值，其卻無法評估參數值。 一般來說，不僅需要瞭解電池的充電狀態，亦需瞭解 電池的健康狀態。於本文中，電力衰退係指電池在老化過 私中，其電阻逐漸增加的現象，且此漸增的電阻使得電池 可被獲取/抑制的電力下降；而電荷容量衰退則係指電池在 10老化過程中，其總電荷容量逐漸減少的現象。上述之電池 電阻與電荷容量皆為隨時間改變的參數。 習知係使用下列各種方法評估電池健康狀態：放電測 試法、化學性質相依法（chemistry_dependent method)、歐 姆測試法（Ohmic test)及部分放電法（partial discharge)。其 15中，放電測試法必須將一充滿電力之充電電池進行完全放 電，以量測出總電荷容量，且執行放電測試時，系統的功 能必需中斷並浪費電池原先儲存之能量。化學性質相依法 包括量測鉛酸電池之極板的腐蝕程度、電解液濃度及電壓 驟降法。歐姆測試法包括電阻測試、傳導性（c〇nductance) 20測試及阻抗測試，此法可與模糊邏輯演算法（fuzzy_1〇gic algorithm)及/或類神經網路相結合。但是，上述方法皆為 侵入性的量測。部分放電法及其他方法則將測試中的電池 與一良好電池或一良好電池的模型互相比較。 1272737 由此可知，目前亟需一可同時評估電池狀態與參數的. 方法。此外，亦需要-毋須中斷系統功能且不會浪費電池 原先儲存之能量的測試，並可普遍應用的方法（其係可應用 於不同類型的電化學電池及不同應用上）、不需侵入㈣測 5的方法及更為精確的方法。且需要一可自動量測諸如電池 電阻及電荷容量等隨時間變化（time參數之方法與 裝置。亦希望能具有一可在串聯及/或並聯等不同設定之電 池組中運作的方法。 10 【發明内容】 本發明±要係提供-評估一電化學電池之目前增量狀 態之方法，此方法包括下列步驟：形成此電化學電池系統 -内部增量狀態預測，其中此增量狀態包括至少一内部狀 怨值及至少-内部參數值；形成此内部增量狀態預測之一 15不確定性預測；校正此内部增量狀態預測及此不確定性預 測；以及執行-演算法，其中此演算法係重複此形成内部 增量狀態預測、此形成此内部增量狀態預測之不確定性預 測及此校正此内部增量狀態預測及此不確定性預測，俾對 此增量狀態產生一現行評估並對此增量狀態評估產生一現 20 行不確定性。 本發明之另-實施例係提供—評估一電池組系統之目 前增量狀態之裝置，其包括：—形成一電池之内部增量狀 態預測的單元；-形成此電池内部增量狀態預測之不破定 性預測的單元；-校正此内部增量狀態預測及此不確定性 1272737 預測的單元；以及一執行一演算法的單元，其中此演算法 係重複由此形成此内部增量狀態預測的單元、此形成此内 部增量狀態預測之不確定性預測的單元及此校正此内部增 量狀態預測及此不確定性預測的單元所執行之步驟，俾^ 5此增量狀態產生-現行評估、並對此增量狀態評估產生一 現行不確定性。 、，本發明之又-實施例係提供一評估一電化學電池之目 前增量狀態之系統’其包括：一對此電化學電池形成一内 部增量狀態預測的機制’其中此增量狀態包括至少一内部 10狀態值及至少-内部參數值；一對此内部增量狀態預_ 成-不確定性預測的機制；一校正此内部增量狀態預測及 此不確，性預測的機制；以及一執行一演算法的機制，其 中此演算法係重複此形成此内部增量狀態預測的機制、此 形成此内部增量狀態預測之不確定性預測的機制及此校正 15 j内部增量狀態預測及此不衫性預測的機制，俾對此增 里狀態產生-現行評估、並對此增量狀態評估產生一現行 一儲存媒體，其係以一機 中此儲存媒體包括一驅使 電池之目前增量狀態的方 本發明之再一實施例係提供 器可讀取之電腦程式碼編碼，其 電月自執行上述之評估一電化學 法0 本發明之再—實施例係提供-藉由電腦資料訊 $之傳«體’其中此電腦資料訊號包括，驅使一電 執仃上述之評估一電化學電池之目前增量狀態的方法。 20 1272737 【實施方式】 ▲於本發日月中所揭露之實施例包括利用聯合遽波法以 評估一電化學電池之狀態及參數的方法、系統及裝置。請 參閱圖1及圖2，本文將於後文中提出許多具體細節俾對本 5發明做更詳盡的解釋。需注意的是，於本文之實施例中所 運用之電池及電化學電池包含但不限於：電池、電池組、 超電容（ultracapacitor)、電容組（capacitor bank)、燃料電池 (fuel cell)、電解電池（electr〇lysis cell)及其他等效電池或 結合上述至少一電池種類而構成之電池。此外，於本文所 10提及之電池或電池組可包括複數個電池，且本文所揭露之 實施例則應用一或多個上述之電池。 本發明之一或多個評估電池狀態及參數值之實施例 係運用聯合遽波法。本文之一或多個評估電池狀態及參數 值之實施例係運用聯合卡爾曼濾波法（j〇int Kalman 15 filtering)。本文之部分評估電池狀態及參數值之實施例係 運用聯合延伸卡爾曼濾波法（j〇int extended Kalman filtering)。本發明之部分實施例同時評估充電狀態⑻攸〇f charge，S0C)、電力衰退（p0wer fade)及/或電荷容量衰退 (capacity fade);而本發明之其他實施例則評估額外的電池 20狀悲值及/或額外的隨時間變化的參數值。此外，本發明之 實施例中所提及之「濾波法」（filtering)係傾向於包括遞迴 (recursive)預測及校正的濾波方法，且包括但不限於卡爾 曼濾波法及/或延伸卡爾曼濾波法。 11 1272737 圖1係顯示本發明一實施例之參數評估系統1〇之元 件。其中，電化學電池組20包括複數個電池22，例如電池 係連接至一負載電路30 (load circuit)。舉例來說，負載電 路30可以是一電動車（eiectric vehicle，EV)或一複合式電動 5 車（hybrid electric vehicle，HEV)之馬達。量測裝置4〇 則係 量測各種電池的特徵及屬性，而此量測裝置4〇包括但不限 於一量測電池之終端電壓（terminal voltage)的裝置，例如 一電壓感測器42，其可以是一伏特計（v〇ltmeter)或其他等 效之感測器，而電流感測器44則量測電池之電流，其可以 10是一安培計（ammeter)或其他等效之感測器。此外，亦可選 擇性地使用溫度感測器46量測電池之溫度，其可以是一溫 度計（thermometer)或其他等效之感測器。而諸如内部壓力 或阻抗（impedance)等其餘的電池特性，則可運用例如壓力 感測器及/或阻抗感測器48加以量測，且可被應用於電池組 15 20之某些種類的電池22中。本實施例可運用各種所需的感 測器來評估電池22的特徵及屬性。前述之電壓、電流、（溫 度）及電池屬性的量測值則由運算電路5〇 ciixuit)處裡，運算電路5〇可以是一處理器或一電腦，其評 估電池22之參數。此系統亦具有一儲存媒體52，其可為任 20何熟知該項技術領域者所知悉之可運用於電腦系統中之儲 存媒體。儲存媒體52可藉由許多種不同方式與運算電路5〇 間建立一溝通連結，例如透過（但不限於）傳播訊號Μ (propagated signal)與運算電路5〇互相溝通。需注意的是， 本發明不需使用任何裝置量測電池之内部化'學:件。此 12 1272737 外，本發明所有使用之量測方法皆為非侵入性 (non-invasive)的量測方法，亦即本發明所使用之量測方法 並不會對受測系統的使用或資料輸入產生任何干擾而影響 到負載電路30的正常運作。 5 為了實現預定之功能及理想之處理過程與運算，例如 模型化、评估預定參數等。運算電路5〇包括但不限於處理 器閘陣列（gate array)、客製化邏輯（cust〇m i〇gic)、電腦、 記憶體、儲存空間（storage)、暫存器（register)、計時器 (timing)、中斷器（interrupt)、溝通介面及輸入/輸出訊號介 10面或上述至少一元件的組合。運算電路50亦可具有輸入器 及輸入訊號濾波法（input signai fiitering)或其他等效的裝 置，俾確保能自溝通介面與輸入器精準地採樣及轉換或取 得訊號。至於運算電路50的額外特徵及特定的處理過程， 將於後文中詳述。 15 本發明之一或多個實施例係以全新或更新之勒體或 軟體的形式實施，其係由運算電路5〇及/或其他處理控制器 加以執行。軟體功能包括但不限於韌體，且可實施於硬體、 軟體或兩者結合之平台上。據此，本發明之一顯著優勢係 在於其可實施於現有及/或全新處理系統中，俾對一電化學 20電池進行充電及控制。 於本發明之一實施例中，運算電路50使用電池22之一 數學模型，此模型具有一動態（dynamic)系統狀態之指標 (indicia)。於本發明之一實施例中，係使用一非連續時間 模型（discrete-time model)。舉例來說，一非連續時間之狀 13 1272737 態空間（state space)的模型（其可能為一非線性模型）具有下 列的表現形式： ^k+i=f(xk^ukA) + wk ^ yk=g(x^ukA) + vk (式υ 其中，Α為系統狀態（system state)，^為隨時間變化之 5 模型參數組（time varying model parameters)，w為外部輸入 (exogenous input) 5 yk% % tB (system output) j v/；為雜訊輸入（noise input)。此外，所有數值皆可以為數值 (scalar)或向量（vector)。/「·，·，.）及g〔·，·，·）為電池模型所使用 之函數，而那些電池模型所需之非隨時間變化之數值 10 (non-time-varying numeric value)則被納入於函數/「·”，」及 gf·，·，·）中，並不包含於心中。 系統狀態至少包括一最小數目的資訊，即預測電池22 之目前輸出所需之電池22目前的輸入及所使用之數學模 型。對電池22而言，系統狀態可能包括：充電狀態、對應 15 於不同時間常數之極化電壓程度（polarization voltage level) 及磁滯程度（hysteresis level)。至於系統之外部輸入w，最 小必需包括目前電池之電流“，且選擇性地包括電池的溫 度（除非溫度變化已經被納入電池之數學模型的系統狀態 中）。由於系統參數心之數值僅隨著時間緩慢地變化，所以 20 系統參數心並無法直接由系統所量測的輸入及輸出決定其 值。系統參數心包括但不限於：電池電荷容量（cell capacity)、電阻（resistance)、極化電壓時間常數 (polarization voltage time constant)、極化電壓混合係數 1272737 (polarization voltage blending factor)、磁滯混合係數 (hysteresis blending factor)、磁滯現象比率常數（hysteresis rate constant)及效率常數（efficiency factor)等。模型輸出少灸 則對應於物理可量測之電池數值（cell quantities)或可直接 5 由量測之電池數值計算而出的數值，其至少包括負載情形 下之電池電壓（cell voltage under load)。 本實施例亦可應用具有動態參數的數學模型，如下所 示： θ^\ =°k+rk (式 2) 10 此方程式顯示出參數本質上是固定不變的，但其卻會 緩慢地隨著時間而改變，而此種現象係由一虛擬之雜訊程 序G代表。 請參閱圖2，於其所示之聯合濾波器100中，狀態之動 態及參數之動態被結合而形成一增量系統（augmented 15 system)，如下列模型所示：1272737 IX. Description of the Invention: [Technical Field] The present invention relates to a method and apparatus for evaluating battery system status and module parameters using digital filtering techniques, and more particularly to using a combined card 5 mann wave method ( Joint Kalman filtering) and joint extension Kalman filtering battery unit system state and module parameter evaluation device and method. [Prior Art] 10 In the technical field of rechargeable battery packs, it is preferable to have an application function that can evaluate and describe the current state of the battery pack. However, in reality, the current state values of some battery packs cannot be directly measured. Moreover, some of the state values change quite rapidly, such as the state of charge (S0C) of the battery pack, and the change range 15 can reach its complete application range in a short time; otherwise, the state value of the part is The change is quite slow, for example, the cell capacity is only about 20% under the premise of normal use for decades. In general, state values that tend to change rapidly include the state of the system, while state values that tend to change slowly include those in the system that change over time. In the technical field of battery systems, such as hybrid electric vehicles (HEVs), battery electric vehicles (BEVs), notebook batteries, portable tool battery packs and other equivalents As long as it works as long as possible and does not harm the 1272737 eight-battery battery system, it is better to use the information related to the rapid change. Most of the information (such as the state of charge parameters) is evaluated. ; available electricity. In addition, it is preferable to confirm the information related to the parameters of the slow change (such as the total charge capacity parameter), to ensure accurate prior calculations and extend the service life of the battery pack during the life cycle of the battery pack. And to help determine the health of the battery pack (state h^lth, SOH) 〇 'There are many ways to evaluate the state of the battery, which are roughly evaluated for the following two state values: state of charge (SOC), power fade And 10 charge capacity decline (capacity fade), wherein the state of charge is a rapidly changing state value, and power decay and charge capacity decline are all state values of slow change k. In addition, if the current and initial battery pack resistances are known, the power decay can be calculated; if the total charge capacity of the current and initial battery packs is known, the charge capacity decay can be calculated. Of course, power decay and charge capacity degradation can also be calculated using other different methods, and are not limited to the above methods. When describing the health status (s〇H) of a battery pack, the power decline is confused with the electric power. Using the above variable values, some other information can be derived, such as the maximum available power of the battery pack at any point in time. In addition, in some specific applications, additional state members or parameters are used, and these state components or parameters are determined using individual calculations. The state of charge (SOC) is a value that is usually expressed as a percentage and indicates the current available power of the battery pack. There are many different methods currently used to evaluate the state of charge of a battery pack, including: discharge 1272737 discharge test, ampere-hour counting (or coulomb counting). , measuring electrolyte, open-circuit voltage measurement, linear and nonlinear circuit modeling, impedance spectroscopy, Internal resistance measurement, voltage dip method (Coup de fouet) and some forms of Kalman filter. Among them, the discharge test method must completely discharge the battery to judge the state of charge, and it is necessary to interrupt the system function and take a long time to perform the discharge test. Therefore, the application of the discharge test is limited. The ampere-hour counting method (or coulomb calculation method) adopts the method of "open loop", and its accuracy is reduced with time due to the accumulation of the measurement error value. The measurement electrolyte method is only applicable to vented lead-acid batteries, and its applicability is rather limited. The open circuit voltage measurement method is only applicable to cells with hysteresis effect and can only be applied if the battery is not used for a long period of time. In addition, open circuit voltage measurement does not work in a dynamic environment. Linear and nonlinear circuit modeling methods cannot directly calculate the state of charge of the battery. The state of charge of the battery must be estimated from the calculated values. The measurement required for impedance spectrum analysis is not as high as 20% for general applications. The internal resistance measurement method is very sensitive to the measurement error value' and the measurement required is not possible in general applications. The voltage dip method is only suitable for use in lead acid batteries. Some forms of Kalman filtering that do not use the state of charge (SOC) as a filter state do not directly calculate the error bound of the measured value. In other methods, for example, U.S. Patent No. 1,272,737 to U.S. Patent No. 6,534,954, the entire disclosure of which is incorporated herein by reference in its entirety, by the use of The current and temperature measurements are evaluated by a filter (preferably Kalman; wave filter) # to estimate the state of charge of the battery. However, although this method 5 can directly evaluate the status value, it cannot evaluate the parameter value. In general, you need to know not only the state of charge of the battery, but also the health of the battery. In this paper, power decay refers to the phenomenon that the battery is aging and its resistance gradually increases, and this increasing resistance causes the battery to be taken/suppressed by the power to decrease; while the charge capacity decline refers to the battery aging at 10 In the process, its total charge capacity gradually decreases. The above battery resistance and charge capacity are parameters that change with time. Conventional methods use the following methods to evaluate battery health: discharge test, chemistry_dependent method, Ohmic test, and partial discharge. In the 15th, the discharge test method must fully discharge a fully charged rechargeable battery to measure the total charge capacity, and when performing the discharge test, the function of the system must be interrupted and the energy originally stored by the battery is wasted. The chemical phase includes the measurement of the corrosion degree of the plate of the lead-acid battery, the electrolyte concentration and the voltage dip method. The ohm test method includes a resistance test, a conductivity (c〇nductance) 20 test, and an impedance test, which can be combined with a fuzzy logic algorithm (fuzzy_1〇gic algorithm) and/or a neural network. However, all of the above methods are invasive measurements. The partial discharge method and other methods compare the battery under test with a model of a good battery or a good battery. 1272737 It can be seen that there is a need for a method for simultaneously evaluating battery status and parameters. In addition, there is a need for - no need to interrupt system functions without wasting the energy stored in the battery, and a universally applicable method (which can be applied to different types of electrochemical cells and different applications) without intrusion (4) 5 methods and more precise methods. There is also a need for a method and apparatus for automatically measuring time-dependent changes such as battery resistance and charge capacity. It is also desirable to have a method that can operate in battery packs of different settings such as series and/or parallel. SUMMARY OF THE INVENTION The present invention provides a method for evaluating the current incremental state of an electrochemical cell, the method comprising the steps of: forming the electrochemical cell system - internal incremental state prediction, wherein the incremental state comprises at least one Internal grievances and at least - internal parameter values; forming one of the internal incremental state predictions 15 uncertainty prediction; correcting this internal incremental state prediction and this uncertainty prediction; and performing-algorithm, where the algorithm Repetitively forming an internal incremental state prediction, this uncertainty prediction that forms this internal incremental state prediction, and correcting this internal incremental state prediction and this uncertainty prediction, generating an ongoing assessment of this incremental state and This incremental state assessment yields a current 20-line uncertainty. Another embodiment of the present invention provides for evaluating the current increase in a battery system. a state device comprising: - a unit that forms an internal incremental state prediction of a battery; - a unit that forms an undetermined prediction of the internal incremental state prediction of the battery; - corrects the internal incremental state prediction and the uncertainty 1272737 a unit for predicting; and a unit for performing an algorithm, wherein the algorithm repeats the unit for forming the internal incremental state prediction, the unit for forming the uncertainty prediction of the internal incremental state prediction, and correcting the The internal incremental state prediction and the steps performed by the unit of the uncertainty prediction, the incremental state generation - the current evaluation, and the current state of the incremental state evaluation produces a current uncertainty. - Embodiments provide a system for evaluating the current incremental state of an electrochemical cell comprising: a mechanism for forming an internal incremental state prediction for the electrochemical cell wherein the incremental state includes at least one internal 10 state value And at least - internal parameter values; a mechanism for predicting the internal incremental state pre-determination - uncertainty; a correction for this internal incremental state prediction and this Indeed, the mechanism of sexual prediction; and a mechanism for performing an algorithm, wherein the algorithm repeats the mechanism for forming the internal incremental state prediction, the mechanism for forming the uncertainty prediction of the internal incremental state prediction, and the like Correcting 15 j internal incremental state prediction and the mechanism of this non-shirt prediction, 俾 this state of increase - current assessment, and the evaluation of this incremental state produces a current storage medium, which is stored in one machine The medium includes a drive for driving the current incremental state of the battery. A further embodiment of the present invention is a computer readable code that can be read by the provider. The electrical month performs the above-described evaluation and an electrochemical method. The system provides - by computer information, the transmission of the text, including the computer data signal, drives an electric device to evaluate the current incremental state of an electrochemical cell. 20 1272737 [Embodiment] ▲ Embodiments disclosed in this issue include methods, systems, and apparatus for utilizing a combined chopping method to evaluate the state and parameters of an electrochemical cell. Please refer to FIG. 1 and FIG. 2, which will be explained in more detail in the following text. It should be noted that the battery and electrochemical battery used in the embodiments herein include, but are not limited to, a battery, a battery pack, an ultracapacitor, a capacitor bank, a fuel cell, and an electrolysis. A battery (electr〇lysis cell) and other equivalent batteries or batteries combined with at least one of the above types of batteries. Moreover, the battery or battery pack referred to herein may comprise a plurality of batteries, and the embodiments disclosed herein employ one or more of the above described batteries. One or more embodiments of the present invention for evaluating battery status and parameter values utilize a combined chopping method. One or more of the examples herein for evaluating battery state and parameter values employ a joint Kalman filtering method (j〇int Kalman 15 filtering). Part of the evaluation of the battery state and parameter values in this paper is the use of joint extended Kalman filtering. Some embodiments of the present invention simultaneously evaluate the state of charge (8) 攸〇f charge, S0C), p0 wer fade, and/or charge fade (capacity fade); while other embodiments of the present invention evaluate additional battery 20 Sorrow value and / or additional parameter values that change over time. Furthermore, the "filtering" referred to in the embodiments of the present invention tends to include recursive prediction and correction filtering methods, including but not limited to Kalman filtering and/or extending Kalman. Filtering method. 11 1272737 Fig. 1 is a diagram showing an element of a parameter evaluation system according to an embodiment of the present invention. The electrochemical battery pack 20 includes a plurality of batteries 22, such as a battery, connected to a load circuit 30. For example, the load circuit 30 can be an electric vehicle (EV) or a hybrid electric vehicle (HEV) motor. The measuring device 4 量 measures the characteristics and attributes of various batteries, and the measuring device 4 〇 includes, but is not limited to, a device for measuring the terminal voltage of the battery, such as a voltage sensor 42 . It can be a voltmeter or other equivalent sensor, while current sensor 44 measures the current of the battery, which can be an ammeter or other equivalent sensor. In addition, temperature sensor 46 is optionally used to measure the temperature of the battery, which may be a thermometer or other equivalent sensor. Other battery characteristics, such as internal pressure or impedance, can be measured using, for example, a pressure sensor and/or impedance sensor 48, and can be applied to certain types of batteries of battery pack 15 20 22 in. This embodiment can utilize various desired sensors to evaluate the characteristics and attributes of the battery 22. The aforementioned voltage, current, (temperature) and battery properties are measured by an arithmetic circuit 5, which may be a processor or a computer that evaluates the parameters of the battery 22. The system also has a storage medium 52 that can be used by any of those skilled in the art to store media in a computer system. The storage medium 52 can establish a communication link with the computing circuit 5 in a number of different manners, such as by communicating with, but not limited to, a propagated signal and an arithmetic circuit 5〇. It should be noted that the present invention does not require any device to measure the internalization of the battery. In addition to the 12 1272737, all the measurement methods used in the present invention are non-invasive measurement methods, that is, the measurement method used in the present invention does not use or input data to the system under test. Any interference is generated that affects the normal operation of the load circuit 30. 5 In order to achieve the predetermined functions and ideal processes and operations, such as modeling, evaluation of predetermined parameters and so on. The arithmetic circuit 5〇 includes but is not limited to a processor gate array, a customization logic (cust〇mi〇gic), a computer, a memory, a storage, a register, and a timer ( Timing), interrupt, communication interface and input/output signal 10 or a combination of at least one of the above. The arithmetic circuit 50 can also have an input device and input signai fiitering or other equivalent means to ensure accurate sampling and conversion or acquisition of signals from the communication interface and the input device. Additional features and specific processing of the arithmetic circuit 50 will be described in detail later. One or more embodiments of the present invention are implemented in the form of new or updated lexicons or software, which are executed by arithmetic circuitry 5 and/or other processing controllers. Software functions include, but are not limited to, firmware, and can be implemented on hard, soft, or a combination of both. Accordingly, one of the significant advantages of the present invention is that it can be implemented in existing and/or new processing systems to charge and control an electrochemical 20 battery. In one embodiment of the invention, the arithmetic circuit 50 uses a mathematical model of the battery 22 having an indicia of dynamic system state. In one embodiment of the invention, a discrete-time model is used. For example, a discontinuous time model of 13 1272737 state space (which may be a nonlinear model) has the following representation: ^k+i=f(xk^ukA) + wk ^ yk =g(x^ukA) + vk (where Α is the system state, ^ is the time varying model parameters, w is the external input 5 yk % % tB (system output) jv/; is a noise input. In addition, all values can be scalar or vector. /"·,·,.) and g[·,· , ·) is a function used by the battery model, and the non-time-varying numeric value required for those battery models is included in the function / "·"," and gf·, ,·), is not included in the heart. The system state includes at least a minimum amount of information, i.e., the current input of the battery 22 required to predict the current output of the battery 22 and the mathematical model used. For battery 22, the system state may include: state of charge, polarization voltage level corresponding to different time constants, and hysteresis level. As for the external input w of the system, the minimum must include the current current of the battery "and optionally the temperature of the battery (unless the temperature change has been incorporated into the system state of the mathematical model of the battery). Since the value of the system parameter is only The time changes slowly, so the 20 system parameters cannot be directly determined by the input and output measured by the system. The system parameters include but are not limited to: battery capacity, resistance, polarization voltage Time constant (polarization voltage time constant), polarization voltage mixing factor 1272737 (polarization voltage blending factor), hysteresis blending factor, hysteresis rate constant, efficiency factor, etc. The model output less moxibustion corresponds to the physically measurable cell quantity or the value directly calculated from the measured battery value, which includes at least the cell voltage under load. This embodiment can also apply the number with dynamic parameters. The model is as follows: θ^\ =°k+rk (Equation 2) 10 This equation shows that the parameters are fixed in nature, but they slowly change over time, and this phenomenon is caused by A virtual noise program G stands for. Referring to Figure 2, in the joint filter 100 shown, the state dynamics and the dynamics of the parameters are combined to form an augmented system, such as the following model. Show:

Xk+\ 本 ^k+\ _ _ ^ _ 丁 yk =s(^k^kA)+vk 其中，為了簡化標號的複雜度，本文會將具有目前狀 態及目前參數之向量以;^表示。 20 於本發明之實施例中，前述之納入系統狀態之動態及 參數之動態的增量模型係應用一聯合濾波法之程序。再次 15 1272737 重申’此聯合濾波法可運用聯合卡爾曼濾波器1 〇〇或聯合延 伸卡爾曼濾波器100。 請參閱表1，其係運用一聯合延伸卡爾曼濾波法之方 法及系統的實施例。其中，其運算程序係藉由將增量狀態 5評估值么設定為真實增量狀態之最佳猜測（best guess)而 被初始化，而此最佳猜測可藉由將頂端（top portion)設定為 五|>G]及將底端（bottom portion)設定為五[/¾]取得。此外， 評估誤差（estimate-error)之共變異數矩陣g (covariance matrix)亦被初始化。 10 表1 :用於狀態及權重更新之聯合延伸卡爾曼濾波器 狀態空間模型（State-space model): 〜+1 f{xk,uk，ek) 本 % Zk+i = ^(Zk^>uk) + ^k+\. 一 ^ _ r 或 y k ^ S k k ^ k) ^ v k yk = 15 其中，Wh及q分別為共變異數矩陣、Σν& Σ/ 獨立、零均值（zero mean)及高斯（Gaussian)雜说的私序 定義： 4-1 初始化（Initialization): 當众=0時，設定 ςXk+\ this ^k+\ _ _ ^ _ □ yk = s(^k^kA)+vk where, in order to simplify the complexity of the label, the vector will have the current state and the current parameter vector; In the embodiment of the present invention, the aforementioned incremental model incorporating the dynamics of the state of the system and the dynamics of the parameters applies a joint filtering method. Again 15 1272737 reiterates that this joint filtering method can use a joint Kalman filter 1 联合 or a joint extended Kalman filter 100. Please refer to Table 1, which is an embodiment of a method and system using a joint extended Kalman filter. Wherein, the arithmetic program is initialized by setting the incremental state 5 evaluation value to the best guess of the true incremental state, and the best guess can be set by setting the top portion to Five|>G] and set the bottom portion to five [/3⁄4]. In addition, the covariance matrix g of the estimate-error is also initialized. 10 Table 1: State-space model for joint extension Kalman filter for state and weight update: ~+1 f{xk,uk,ek) %% Zk+i = ^(Zk^> Uk) + ^k+\. a ^ _ r or yk ^ S kk ^ k) ^ vk yk = 15 where Wh and q are the covariance matrix, Σν& Σ/ independence, zero mean and Gauss (Gaussian) miscellaneous private order definition: 4-1 Initialization: When public = 0, set ς

Xk-\=Xk-\ uk) ^XkXk-\=Xk-\ uk) ^Xk

Xk^k 20 1272737 Z〇+ = Ε[χ0],ΣΙ〇 = Ε[(χ0 - f0+)(- f〇+)Γ] 運算（computation): 當众=1、2、…時，計算： 時間更新（time update)Xk^k 20 1272737 Z〇+ = Ε[χ0], ΣΙ〇= Ε[(χ0 - f0+)(- f〇+)Γ] operation: When public = 1, 2, ..., calculate: time Update (time update)

Zk = F(zU，uk^) 5 = Ak_x ATk i +diag(Tw^r) 量測更新（measurement update)Zk = F(zU, uk^) 5 = Ak_x ATk i +diag(Tw^r) Measurement update

Lk 喝，“C孤 Σ，)7^]-1 χ\ = Xl + Lk[yk + g(tk^k)] =(I-LkCk)Y}>k ------^__ 於此範例中，在每一量測間隔中，許多步驟被執行。 10首先，增量狀態評估值f (augmented state estimate)係透過 函數F卩通著時間改變，增量狀態向量之不確定性亦同時被 更新。上述之表1僅提供不確定性評估（uncertainty estimate) 之更新方式的其中一個範例，其實際上仍有許多其他可能 的更新方式。電池輸出之量測值係與由增量狀態評估值f 15產生之預測輸出互相比較，而其間的誤差值則用於更新增 里狀恶評估值i的值。此外，表1所敘述的步驟可依照各種 不同順序加以執行，並非以表1所列出之執行順序為限，任 何^知該項技術領域者將可推導出各種具有相同功效之順 序組合的運算式。 2〇 清繼續參閱圖2，其係描述本發明之一實施例的實施 例。單一據波器100係同時更新狀態及參數評估。此濾波 17 1272737 器具有一時間更新或預測101單元及一量測更新或校正ι〇2 單元。時間更新/預測單元1〇1係接收先前之外部輸入…7 (例如電池之電流及/或溫度）、併同先前評估之增量狀態值 ^及先前評估之增量狀態不確定性評估值益知做為輸入 5 (input)。時間更新/預測單元1〇1提供預測之增量狀雜一及 預測之增量狀態不確定性^；#並輸出至增量狀態量測更新/ 校正單元102。狀態量測更新/校正單元1〇2亦可於提供目前 之增量狀態評估值芯及增量狀態不確定性評估值ς“時，接 收預測之增量狀態芯、預測之增量狀態不確定性、外部 10輸入W、及系統輸出八(system output)。其中，負號標記㈠ 係代表此向量係來自濾波器100之預測元件1〇1的結果；而 正號標記（+)則代表此向量係來自濾波器1〇〇之校正元件 102的結果。 本發明所提出之數個實施例皆須利用一電池狀態的 15數學模型及某些特殊應用之輸出動態（output dynamics)， 而上述之範例係藉由對一般函數介.，一及#·, . ·)進行特殊 定義而達成。 本發明之一實施例係使用一電池模型，其包括電池22 之一或多個開路電壓（open circuit voltage，0CV)、内部電 20 阻（internal resistance)、電壓極化時間常數及磁滯程度。同 理’參數值包括但不限於：效率常數，例如庫侖效率 (Coulombic efficiency)，係標記為 ^ ;電池電容（Cell capacity)，係標記為Ca ;極化電壓時間常數，係標記為 18 1272737 au··'/’極化電壓混合係數，係標記為& u ;電池電阻， 係標記為A ;磁滯混合係數，係標記為;磁滯現象比率 常數’係標記為匕。此外，參數值亦可包括上述至少一種 參數值之組合。為了範例之目的，上述之參數值被調整以 5滿足此模型結構，使此模型可模型化一高功率之鋰高分子 電池（Lithium-ion Polymer Battery，LiPB)的動態，即便目前 提出之結構與方法係為通用形式並應用於其餘之電化學 (electrochemistry)領域。 於本發明之一實施例中，充電狀態係自模型之一狀態 10 擷取出。以下方程式係表示充電狀態： zk+x^zk-irli^tlCk)ik (式 3) 其中’ △(係為採樣間週期（inter·sample period)(以秒計 算），Q係表示電池電容（單位為安培_秒），以係為電池22在 時間指數灸時的充電狀態；k係為電池22之電流；而&係為 15 電池22在電流程度為“時之庫侖效率。 於另一實施例中，極化電壓程度係自多個濾波器狀態 中擷取出。倘若假設共有η/ί固極化電壓時間常數，則 Λ+ι = ^ffk + Bfh (式 4) 其中’矩陣冷e 9iW/XW/可以是一由極化電壓時間常數 20 au···%，*構成之具有實數值（real-valued)的對角矩陣 (diagonal matrix)。如此的話，則當系統於所有輸入（emry) 的強度（magnitude)皆小於1時，此系統呈現穩定狀態 (stable)。此時，向量便可藉由設定"/為1而簡化。 1272737 向量心之輸人並不料，只要所輸人的數值不為零即可。_ 々矩陣之^^的輸人值係選自部分系統確認程序，以適切地 使核型參數符合於量測之電池資料。々及鱗陣在正常電 池組的運作情形下，會隨著時間及其他因子而改變。 於又一實施例中，磁滯程度係自單一狀態 fLk drink, "C orphan," 7^]-1 χ\ = Xl + Lk[yk + g(tk^k)] =(I-LkCk)Y}>k ------^__ In this example, in each measurement interval, many steps are performed. 10 First, the incremental state estimate f (augmented state estimate) is changed by the function F卩, and the uncertainty of the incremental state vector is also It is also updated at the same time. Table 1 above provides only one example of how to update the uncertainty estimate. There are actually many other possible updates. The battery output is measured by the incremental status. The predicted outputs produced by the evaluation value f 15 are compared with each other, and the error values therebetween are used to update the value of the increased value of the evaluation value i. Furthermore, the steps described in Table 1 can be performed in various orders, not in Table 1. The order of execution listed is limited, and any one skilled in the art will be able to deduce various arithmetic expressions having the same order of functions. 2 Continuation Referring to Figure 2, an embodiment of the present invention is described. Embodiment. A single volute 100 is capable of simultaneously updating status and parameter evaluation. 17 1272737 has a time update or prediction 101 unit and a measurement update or correction ι〇2 unit. The time update/prediction unit 1〇1 receives the previous external input...7 (eg battery current and/or temperature), and The incremental state value ^ and the previously evaluated incremental state uncertainty evaluation value are used as input 5 (input). The time update/prediction unit 1〇1 provides the predicted increment and prediction. The incremental state uncertainty ^;# is output to the incremental state measurement update/correction unit 102. The state measurement update/correction unit 1〇2 can also provide the current incremental state evaluation value core and the incremental state is not The deterministic evaluation value ς "when the predicted incremental state core is received, the predicted incremental state uncertainty, the external 10 input W, and the system output VIII (system output). The negative sign (1) represents the result of the vector from the prediction component 1〇1 of the filter 100; and the positive sign (+) represents the result of the vector from the correction component 102 of the filter 1〇〇. Several embodiments of the present invention are required to utilize a mathematical model of a battery state of 15 and output dynamics of certain special applications, and the above examples are based on general functions, one and #. ·) Achieve a special definition. One embodiment of the present invention utilizes a battery model that includes one or more of open circuit voltage (0 CV), internal electrical resistance, voltage polarization time constant, and hysteresis. Similarly, the parameter values include, but are not limited to, efficiency constants, such as Coulombic efficiency, labeled as ^; Cell capacity, labeled Ca; polarization time constant, labeled 18 1272737 au ··//'s polarization voltage mixing coefficient, labeled &u; battery resistance, labeled A; hysteresis mixing coefficient, marked as; hysteresis ratio constant constant 'marked as 匕. Furthermore, the parameter value may also comprise a combination of at least one of the above parameter values. For the purposes of the example, the above parameter values are adjusted to 5 to satisfy the model structure, so that the model can model the dynamics of a high-power Lithium-ion Polymer Battery (LiPB), even if the proposed structure and The method is in a general form and is applied to the rest of the field of electrochemistry. In one embodiment of the invention, the state of charge is taken from one of the states of the model. The following equation indicates the state of charge: zk+x^zk-irli^tlCk)ik (Equation 3) where '△( is the inter-sample period (in seconds), Q is the battery capacitance (unit It is ampere-seconds, which is the state of charge of battery 22 during time index moxibustion; k is the current of battery 22; and & is 15 of battery 22 at the current level of "coulomb efficiency." In the example, the degree of polarization voltage is extracted from multiple filter states. If assuming a total η/ί solid polarization voltage time constant, then Λ+ι = ^ffk + Bfh (Equation 4) where 'matrix cold e 9iW /XW/ can be a diagonal matrix of real-valued consisting of a polarization voltage time constant of 20 au···%,*. In this case, when the system is at all inputs (emry When the magnitude of the magnitude is less than 1, the system assumes a stable state. At this time, the vector can be simplified by setting "/ to 1. 1272737 The input of the vector heart is unexpected, as long as the person is input The value of the value is not zero. The input value of the ^^ matrix is selected from the partial system confirmation. In order to properly conform the karyotype parameters to the measured battery data, the squama and scale arrays may change over time and other factors in the normal battery pack operation. In yet another embodiment, the degree of hysteresis From a single state f

hk+x = exP -VHk+x = exP -V

\khrk^ \ f f ck K + / 1-exp V V\khrk^ \ f f ck K + / 1-exp V V

VukhYk^t 5 10 中所擷取出 (式5) sgn(4) 其中，h係為磁滯現象比率常數，其亦選自系統確認 程序。 於再一實施例中，整體模型狀態如下 (式6) xk=[fk K Zkf 其狀態亦可依照其他的順序排列。於本例中，模型之 狀態方程式係藉由結合所有前述之個別方程式而形成。 結合狀態值所形成預測電池電壓之輸出方程式如下·· Vk - ^〇CV(zk) + Gkfk^Rkik+Mkhk (式 7 ) 15 其中，G41XB/係為極化電壓混合係成之向 量，其將極化電壓狀態摻合於輸出；Α係為電池電阻（亦可 使用不同的數值表示放電/充電而^^係為磁滯混合係 數。需注意的是，有可能受到設定限制、而使得自&至 的直流增益（DC gain)為零。 2〇 於此範例中’參數係為 - Jh，k，Ck，alk···%，" g'，k.“gn「'，k，γ” Rk，Mk_T (式 g) 20 1272737 增量狀態向量I係藉由將狀態向量（如方程式（6)之狀_ 態向量）及參數向量（如方程式（7)之參數向量）結合而形 成，例如：Taken from VukhYk^t 5 10 (Equation 5) sgn(4) where h is the hysteresis ratio constant, which is also selected from the system validation procedure. In still another embodiment, the overall model state is as follows (Equation 6) xk = [fk K Zkf The states may also be arranged in other orders. In this example, the equation of state of the model is formed by combining all of the aforementioned individual equations. The output equation for predicting the battery voltage formed by combining the state values is as follows: · Vk - ^ 〇 CV(zk) + Gkfk^Rkik + Mkhk (Expression 7) 15 where G41XB/ is a vector of polarization voltage mixing system, which will The polarization voltage state is blended into the output; the lanthanum is the battery resistance (other values can be used to indicate discharge/charge and ^^ is the hysteresis mixing coefficient. It should be noted that it may be subject to setting restrictions, and The DC gain is zero. 2 In this example, the 'parameters are - Jh,k,Ck,alk···%," g',k."gn"',k,γ Rk, Mk_T (Formula g) 20 1272737 The incremental state vector I is formed by combining a state vector (such as the state _ state vector of equation (6)) and a parameter vector (such as the parameter vector of equation (7)). E.g:

Xk=lXk, K Ck, au...anf k, ghk...gn^]k^ Mjr 5 10 15 其中，增量狀態向量中之狀態及參數亦可依其他的順 序排列。心中之數值係包括所有計算函數介八例如方程 式（3)至（5))及g「·,·，·」（例如方程式（7))之細節。 於任一實施例中，聯合濾波器100係適應於一狀態評 估及一參數評估，俾使模型之輸入-輸出關係與量測到之輸 入-輸出資料儘可能地相符。但是，並不保證模型之增量狀 態會收斂至物理增量狀態值。於一實施例中，聯合淚波法 所使用之電池模型尚可能藉由加入一第二電池模型進行擴 充增補，其中，第二電池模型係具有使得增量狀態收斂於 正確值之輸出。一實施範例係執行額外的步驟以確保一模 型之增量狀態收斂於充電狀態（SOC): g{xk^uk^0)= ocv(zk) - RjA + K + Gkfk zk (式9) 增補之模型輸出係與聯合濾波器100之量測輸出互相 比較。於一實施例中，充電狀態之量測值可能趨近於Xk=lXk, K Ck, au...anf k, ghk...gn^]k^ Mjr 5 10 15 Among them, the state and parameters in the incremental state vector can also be arranged in other order. The values in the mind include details of all calculation functions, such as equations (3) through (5)) and g "·, ·, ·" (for example, equation (7)). In either embodiment, the joint filter 100 is adapted to a state evaluation and a parameter evaluation such that the input-output relationship of the model matches the measured input-output data as much as possible. However, there is no guarantee that the incremental state of the model will converge to the physical incremental state value. In one embodiment, the battery model used in conjunction with the tear wave method may be expanded and complemented by the addition of a second battery model having an output that causes the incremental state to converge to the correct value. An embodiment performs additional steps to ensure that the incremental state of a model converges to a state of charge (SOC): g{xk^uk^0) = ocv(zk) - RjA + K + Gkfk zk (Equation 9) Supplement The model output is compared to the measured output of the joint filter 100. In an embodiment, the measured state of the charging state may approach

Lk 其係由下列方程式推導出 ykK〇CV(zk) 一 Rkik 〇CV(zk)^yk+Rkik (式 1〇) zk=〇CV-\vk+Rkik) 21 20 1272737 藉由量測負載情形下之電池的電壓及電流’具備對兄 的知識〆或經由聯合濾波器1〇〇之么）、以及瞭解電池化學之 反向開路電壓（inverse OCV)的功能，本實施例可計算出充 電狀態之雜訊評估值（noisy estimate)之。 於此實施例中，一聯合濾波器100執行於此校正之模 型中，於量測更新中具有量測資訊，即Lk is derived from the following equation: ykK〇CV(zk) - Rkik 〇CV(zk)^yk+Rkik (formula 1〇) zk=〇CV-\vk+Rkik) 21 20 1272737 by measuring the load case The battery voltage and current 'has the knowledge of the brother 〆 or via the joint filter 1 、), and the function of understanding the reverse OCV of the battery chemistry, this embodiment can calculate the state of charge Noise evaluation value (noisy estimate). In this embodiment, a joint filter 100 performs the calibration model and has measurement information in the measurement update, that is,

Lzd 實驗證明，當&之雜訊（由磁滯效應造成之短期誤差及 被忽略之極化電壓）被禁止做為充電狀態之主要量測工具 10時，其於動態環境中之長期行為的預測將會是準確的，且 維持聯合濾波器100中之充電狀態的精準度。 练上所陳，前述各實施例係描述一同時進行電池狀態 及參數狀態評估之方法。其中之一或多個實施例係使用一 卡爾叉濾波器100,而部分實施例係使用一延伸卡爾曼濾波 15器100此外彳为實施例係具有一驅使充電狀態趨於收敛 的機制。本發明具有廣泛地應用範圍及可應用於電池電化 學領域中。 t 發明所揭露之方法可以電腦可實現之程序及實施 這些程序之裝置的形式實施。本方法亦可為一包:有t 20之電腦程式碼的形式，俾嵌入於實體媒體52中。實體媒體 =:ΓΓ:0Μ、硬碟機、或其他任何電腦可讀取 中：m 腦程式碼被載入並被執行於-電腦 來，此電腦即成為一可執行本發明之方法的裝 22 1272737 置。本方法亦可為肷入於電腦程式碼的形式，舉例來說， 不論儲存於一記錄媒體中、載入及/或執行於一電腦中或為 透過傳輸媒體傳遞之資料訊號54。其中，資料訊^虎54$@ 一調變載波（modulated cairier)或其他種類的訊號，而傳輸 媒體可為電線（electrical wiring)、電纜（cabling)、光纖（仙以 optics)或電磁輻射（electromagnetic radiation)。而當電腦程 式碼被載入並被執行於一電腦中時，此電腦即成為一可執 行本發明方法之裝置。當此電腦程式碼實現於一微處理器 ίο 時’此電腦程式碼將設定此微處理器以建立一特定邏輯電 路（logic circuit) 〇 上述實施例僅係為了方便說明而舉例而已，本發明所 主張之權利範圍自應以申請專利範圍所述為準，而非僅限 於上述實施例。 15 【圖式簡單說明】 圖1係本發明一較佳實施例執行狀態及參數評估之系統之 示意圖；以及 圖2係本發明一較佳實施例之聯合濾波器之示意圖。 20【主要元件符號說明】 22電池 42電流感測器 50運算電路 52儲存媒體 10參數評估系統 20電池組 3〇負載電路 40量測裝置 44電壓感測器 46溫度感測器 48壓力感測器及/或阻抗感測器 23 1272737 聯合濾波器 54傳播訊號 100 101時間更新/預測單元 102量測更新/校正單元 24The Lzd experiment proves that when & noise (short-term error caused by hysteresis effect and neglected polarization voltage) is prohibited as the main measurement tool 10 for charging state, its long-term behavior in dynamic environment The prediction will be accurate and maintain the accuracy of the state of charge in the joint filter 100. As described above, the foregoing embodiments describe a method for simultaneously performing battery state and parameter state evaluation. One or more of the embodiments use a Karl fork filter 100, and some embodiments use an extended Kalman filter 100. In addition, the embodiment has a mechanism for driving the state of charge to converge. The invention has a wide range of applications and is applicable to the field of battery chemistry. The method disclosed by the invention can be implemented in the form of a computer achievable program and a device for implementing the program. The method can also be a package: in the form of a computer code of t20, embedded in the physical medium 52. Physical media =: ΓΓ: 0 Μ, hard drive, or any other computer readable: m brain code is loaded and executed on the computer, this computer becomes a device that can perform the method of the present invention 22 1272737 set. The method may also be in the form of a computer code, for example, stored in a recording medium, loaded and/or executed in a computer or transmitted through a transmission medium. Among them, the information is ^54$@modulated cairier or other kinds of signals, and the transmission medium can be electrical wiring, cabling, optical fiber (optics) or electromagnetic radiation (electromagnetic Radiation). When the computer program code is loaded and executed in a computer, the computer becomes a device that can perform the method of the present invention. When the computer program code is implemented in a microprocessor ίο 'this computer program code will set the microprocessor to establish a specific logic circuit. The above embodiments are merely examples for convenience of description, the present invention The scope of the claims is subject to the scope of the patent application and is not limited to the above embodiments. 15 is a schematic diagram of a system for performing state and parameter evaluation according to a preferred embodiment of the present invention; and FIG. 2 is a schematic diagram of a joint filter according to a preferred embodiment of the present invention. 20 [Main component symbol description] 22 battery 42 current sensor 50 arithmetic circuit 52 storage medium 10 parameter evaluation system 20 battery pack 3 load circuit 40 measuring device 44 voltage sensor 46 temperature sensor 48 pressure sensor And/or impedance sensor 23 1272737 joint filter 54 propagation signal 100 101 time update/prediction unit 102 measurement update/correction unit 24

Claims (1)

1272737 十、申請專利範圍： 1 · 一種評估一電化學電池系統之目前增量狀態的方 法，包括下列步驟： 形成该電化學電池系統一内部增量狀態預測，其中該 5增量狀態包括至少一内部狀態值及至少一内部參數值； 形成該内部增量狀態預測之一不確定性預測； 校正該内部增量狀態預測及該不確定性預測；以及 執行一演算法，其中該演算法係重複該形成内部增量 狀態預測、該形成該内部增量狀態預測之不確定性預測及 10 =校正該内部增量狀態預測及該不確定性預測，俾對該增 里狀恶產生一現行評估並對該增量狀態評估產生一 確定性。 个 其中該執行- 15 20 2·如申請專利範圍第1項所述之方法 内。卩增置狀態預測之步驟包括·· 判斷一電流量測值； 判斷一電壓量測值；以及 將該電流量測值及該電壓量測值 中，執行該内部增量狀態預測。 ； 學模 3_如申請專利範圍第2項所述之 不確定性預測之步驟包括： + 該執行 將該電流量測值及該電壓量 中，執行該不確定性預測。 ；一數學模 4.如申請專利範圍第3項所述 驟包括： 方法，其中該校正 25 I272737 計算一增益係數； =該=數、_量測值及該内部增量狀態預 ]计异一杈正内部增量狀態預測，·以及 5 10 15 20 利用該增益係數及該不確定性 定性預測。 杈正不確 ” t//請專4項所述之方法，其中該執行-馮异法之步驟包括： 利賴校正内部增量狀態_及該校正不 剩，獲知該演算法於下一時段重 預 了权更復貫轭時所需的預測。 6·如申請專利範圍第5項所述之方法，其中該演算法 ，至>、-選自於—由—卡㈣濾波器及―延 裔構成之群組。 兩又愿波 =如中請專利範圍第6項所述之方法，其 恕包括至少一選自於一由一充電狀態、一電壓極化^狀 -磁滯程度、-電阻、—電荷容量、_極化電壓時 :電壓混合係數、一磁滞混合係數、一磁滯現、 书數及一效率常數構成之群組。 半 8.如申請專利範圍第2項所述之方法，其中 — 内部增量狀態預測之步驟更包括： 量測一溫度；以及 將該溫度量測值、該電流量測值及該電壓量測 於-數學模型中’執行該内部増量狀態預測。 9·如申請專利範圍第8項所述之方法，其中 — 不確定性預測之步驟包括： 7 26 1272737 電壓量測值套用· ’其中該校正步 、一將該溫度量測值、該電流量測值及該 5數于模型中，執行該不確定性預測。 1〇.如申請專利範圍第9項所述之方法 驟包括： / 5 10 15 20 5十算一增益係數； 利用該增益係數、該電壓量測俏 計算一於X hi。， 里劂值及该内部狀態預測， 奴正内部狀態預測；以及 利用該增益絲及料確Μ 定性預測。 t π 杈正不石 U·如申請專利範圍第10項所述 演算法之步驟包括： W之方法，其中該執行- 知校正㈣狀態預測及該校正不確定性預測，豕 …、汁法於下一時段重複實施時所需的預測。 12·如申請專利範圍第U項所述 係至少-選自於一由一卡 〃以次异〉; 器構成之群•且。 ，七及-延伸卡爾曼據》 申4專利範圍第12項所述之古土 能~ # π , 77返之方法，其中該增i 一磁滯φ 兄冤狀恕、一電壓極化程肩 磁尔牲度、一電阻、一電荷交晷 t 了谷里、一極化電壓時間當 電=合係數、一磁滞混合係數、-磁滞現㈣ 吊數及一效率常數構成之群組。 之方法，其中該執行一 14 ·如申晴專利範圍第2項所述 不確定性預測之步驟更包括： 量測一溫度；以及 27 !272737 、w皿度1測值、該電流量 於—數與y讲丨A 則值及該電壓景、日, 數予杈型中，執行該不確定性預測。 里剛值套用 15 ·如申睛專利範圍第丨項所述方 不確定性預測之步驟包括： /’其中該執行一 5 判斷一電流量測值； 判斷一電壓量測值；以及 將該電流量測值及該電壓量測值套 ，執行該内部增量狀態預測。 ；數學模型 16_如申請專利範圍第15項所述之方 1〇不確定性預測之步驟更包括： ，，其中該執行一 量測一溫度；以及 將該溫度量測值、該電流量測值及該電壓旦、， 於一數學模型中，執行該不確定性預測。壓里剛值套用 如申請專利範圍第16項所述之 15 驟包括： /、中该校正步 計算一增益係數； 量狀態預 利用―該增益係數、該電壓量測值及該内部增 測，計算一校正内部狀態預測；以及 20 利用該增益係數及該不確定性預 定性預測。 測，計算一校正不確 18·如申請專利範圍第17項所述之方法，其中該執行一 演异法之步驟包括： 利用該校正内部狀態預測及該校正不確定性預測，獲 知該演算法於下一時段重複實施時所需的預測。 28 1272737 .如巾請專職圍第18項所述之方法，其中該演算法. =-選自於一由一卡爾曼據波器及一延伸卡爾曼滤波 裔構成之群組。 5 '如申請專利範圍第i項所述之方法，其中該方法更 或多增量狀態分別收斂至各自對應之物理數 值的步驟 括提此中請專利範圍第2G項所述之方法，其中該步驟包 10 2;—Λ有—理想收敛值之增量狀態的電池模型輸出。 包括提供 利範圍第21項所述之方法，其中該步驟更 量。A H基於目前量測值之增量狀態的量測向 23.如申請專利範圍第22項所述之方法，其中更包括一 15 電池r:卡爾曼滤波器或一延伸卡爾曼濾波器，利用該 ,q輸出及該量測向量調節一增量狀態評估。 括：24.-種評估-電池組系統之目前增量狀態的裝置，包 二形成一電池之内部增量狀態預測的單元； 20元；< 乂電池内增篁狀悲預測之不確定性預測的單 元；2正該㈣增量狀態預測及該不衫性預測的單 、 '厂、力-/¾ wv早疋，其 成該内部增量狀態預測的單元了 執行-演算法的單元，其中該演算法係重複由該形 該形成該内部增量狀態預 29 1272737 二不墟，疋14預测的單几及該校正該内部增量狀態預測及 H疋性預測的單元所執行之步驟，俾對該增量狀態產 見仃評估、並對該增量狀態評估產生—現行不確定性。 5 電、、也Γ:申請專利範圍第24項所述之裝置，其中該形成-冤池之内部狀態預測的單元包括·· 一量測一電流的單元； 一量測一電壓的單元；以及 二該參數評估、該電流量測值及該電麼量測值，套 10 、學模型中以執行該内部增量狀態預測的單元。 26.如申請專利範圍第25項所述之裝置，其中該 電池二部增量狀態預測之不確定性預測的單元包括:… 中以：將該電流量測值及該電壓量測值套用於-數學模型 中以執行該不確定性預測的單元。 、 請專利範圍第26項所述之裝置，其中該校正該 内。卩增篁狀態預測及該不確定性預測的單元包括： 一計算一增益係數的單元； -利用該增益係數、該電壓量測值及該内 測，計算一校正内部增量狀態預測的的單元；以及恶 利用該增益係數及該不確定性預測，呼筲 20確定性預測的單元。 。十异一校正不 28.如申請專利範圍第27項所述之裝 演算法的單元包括： -卜亥執行該 30 1272737 二利用該&正内部增量狀態制及該校正不確定性預 獲知該演算法於下—時段錢實施時所需之預測的單 ,•如中晴專利範m第28項所述之裝置，其中該演算法 係至>、-選自於一由一卡爾曼滤波器及 器構成之群組。 人應故 10 30.如申請專利範圍第29項所述之袭置，其中該增量狀 :包括，少-選自於一由一充電狀態、一電壓極化程度、 、帶私度電阻、一電荷容量、一極化電壓時間常數、 :極化電壓混合係數…磁滯混合係數、—磁滯現象比率 常數及一效率常數構成之群組。 31.如申請專利範圍第25項所述之裝置，其中該形成該 電池之内部增量狀態預測的單元更包括·· 一量測一溫度的單元；以及 15 一將該溫度量測值、該電流量測值及該電壓量測值套 用於一數學模型中以執行該内部狀態預測的單元。 、32·如中請專利範圍第31項所述之裝置，其中該形成該 電池内部增量狀態預測之不確定性預測的單元包括： 20 測 元 一將該溫度量測值、該電流量測值及該電壓量測值套 用於一數學模型中以執行該不確定性預測的單元。 33_如申請專利範圍第32項所述之裝置，其中該校正該 内部增量狀態預測及該不確定性預測的單元包括： 一計算一增益係數的單元； 31 1272737 -利用該增益係數、該電壓量測值及該内部狀 '、，叶鼻一校正内部增量狀態預測的單元；以及 丨。 確定係數及該不確定性預測以計算-校正不 5 10 34. 如中請專利範圍第33項所述之裝置，其中該執行該 次法的單元包括·· ，：利用該校正内部增量狀態預測、校正内部狀態預測 及忒杈正不確定性預測，獲知該演算法於下一時段 施時所需之預測的單元。 、 係 少-選自於-由-卡爾曼滤波器及一延伸卡爾曼滤油 35. 如申請專利範圍第34項所述之裝置，其中該演算法 器構成之群組 36·如申請專利範圍第35項所述之裝置，其中該增量狀 態包括至少一選自於一由一充電狀態、一電壓極化^度、 15磁滯程度、一電阻、—電荷容量、一極化電壓時間常數、 ，極化電壓混合係數、一磁滯混合係數、一磁滞現象比率 常數及一效率常數構成之群組。 37·如申請專利範圍第25項所述之裝置，其中該形成該 電池内邛增置狀態預測之不確定性預測的單元更包括·· 20 一置測一溫度的單元；以及 一將該溫度量測值、該電流量測值及該電壓量測值套 用於一數學模型中以執行該不確定性預測的單元。 38.如申請專利範圍第24項所述之裝置，其中該形成該 電池内部增量狀態預測之不確定性預測的單元包括： 32 ^72737 一量測一電流的單元； 量測一電壓的單元；以及 5 將該電流量測值及該電壓量 中以執行該不確定性預測。 ’則值套用於一數學模型 10 39.如申請專利範圍第38項所述之裝置 電池内部增量狀態預測之不確定性預測的單元更包括：" 一量測一溫度的單元；以及套用Γ㈣該溫度量測值、該電流量測值及該電壓量測值 '數學杈型中，執行該不確定性預測。 内部4Λ如/請專利範圍第39項所述之裝置，其中該校正該 内4增讀‘4關及該不確定性制的單元包括： 一計算一增益係數的單元； 15 n一利！該增益係數、該電壓量測值及該内部增量狀態 凋，計算一校正内部狀態預測的單元丨以及 一元件利用該增益係數及該不確定性預測，計算一校 正不確定性預測的單元。 41·如申請專利範圍第4〇項所述之裝 演算法的單元包括： 置’其中該執行該 • 利用°亥枚正内部增量狀態預測、校正内部狀態預測 2〇及該校正不確定性預測，獲知該演算法於下一時段’重複實 施時所需之預測的單元。 y 42·如申請專利範圍第41項所述之裝置，其中該演算法 2至少一選自於一由一卡爾曼濾波器及一延伸卡爾曼濾波 器構成之群組。 33 I272737 43 ·如申請專利範圍第24項所述之裝置，其更包括一石崔 保该一或多增量狀態分別收斂至各自對應之物理數值的單 7^ 〇 44.如申請專利範圍第43項所述之裝置，其中該確保該 5 或多增量狀態分別收斂至各自對應之物理數值的單元包 括一提供一具有一理想收斂值之增量狀態之電池模型輸出 的單元。 45·如申請專利範圍第44項所述之裝置，其中該確保該 —或多增量狀態分別收斂至各自對應之物理數值的單元更 1〇包括一提供該一或多增量狀態之基於目前量測值之對應量 測向量的單元。 46. 如申請專利範圍第45項所述之裝置，其中更包括一 卡爾曼濾波器或一延伸卡爾曼濾波器’利用該電池模型輸 出及該量測向量調節一增量狀態評估。 15 47. —種評估一電化學電池之目前增量狀態之系 括： 、匕 一對該電化學電池形成一内部增量狀態預測的裝置， 其中該增量狀態包括至少一内部狀態值及至少一内部來 20值； 口多歎 置一對該内部增量狀態預測形成一不確定性預測的袭 一杈正該内部增量狀態預測及該不確定性預 置；以及 啊展 1272737 一執行一演算法的裝置，其中該演算法係重複該形成 該内部增量狀態的預測、該形成該内部增量狀態預測之不 確定性的預測及該校正該内部增量狀態預測及該不確定性 預測，俾對該肖量狀態產生一現行評估、並對該增量狀態 5 評估產生一現行不確定性。 48· —種儲存媒體，其係以一機器可讀取之電腦程式碼 、烏馬其中忒儲存媒體包括一驅使一電腦執行一評估一電 化學電池之目前增量狀態之方法的指令，該方法包括： 對該電化學電池系統形成一内部增量散態預測，其中 10該增量狀態包括至少-内部狀態值及至少一内部參數值； 對該内部增量狀態預測形成一不確定性預測； 執行一演算法， 法，其中a亥演算法係重複該内部增量狀態1272737 X. Patent Application Range: 1 · A method for evaluating the current incremental state of an electrochemical battery system, comprising the steps of: forming an internal incremental state prediction of the electrochemical battery system, wherein the 5 incremental state comprises at least one An internal state value and at least one internal parameter value; forming an uncertainty prediction of the internal incremental state prediction; correcting the internal incremental state prediction and the uncertainty prediction; and performing an algorithm, wherein the algorithm is repeated Forming an internal incremental state prediction, the uncertainty prediction for forming the internal incremental state prediction, and 10 = correcting the internal incremental state prediction and the uncertainty prediction, generating an current assessment of the increased likelihood and This incremental status assessment produces a certainty. Where the implementation is - 15 20 2 as described in the method of claim 1 of the scope of the patent. The step of predicting the state of the 卩 addition includes: determining a current measurement value; determining a voltage measurement value; and performing the internal incremental state prediction in the current measurement value and the voltage measurement value. ; Learning Mode 3_ The steps of uncertainty prediction as described in item 2 of the patent application scope include: + The execution The current measurement value and the voltage amount are used to perform the uncertainty prediction. A mathematical model 4. The method as recited in claim 3 includes: a method, wherein the correction 25 I272737 calculates a gain coefficient; = the = number, the _ measured value, and the internal incremental state pre-count杈 Positive internal incremental state prediction, and 5 10 15 20 use the gain coefficient and the uncertainty qualitative prediction.杈正不”"//Please refer to the method described in item 4, wherein the execution-Frequency method includes: Relying the internal incremental state _ and the correction is not left, knowing that the algorithm is re-predicted in the next period The predictions required for the weight to suffocate the yoke. 6. The method of claim 5, wherein the algorithm, to >, is selected from the - by-card (four) filter and - A group of constituents. The method of claim 6, wherein the method of claim 6 includes at least one selected from the group consisting of a state of charge, a voltage polarization, a degree of hysteresis, and a resistance. - charge capacity, _polarization voltage: a group of voltage mixing coefficients, a hysteresis mixing coefficient, a hysteresis, a book number, and an efficiency constant. Half 8. As described in claim 2 The method, wherein the step of internally incrementing state prediction further comprises: measuring a temperature; and determining the temperature measurement, the current measurement, and the voltage measurement in a mathematical model to perform the internal measurement state prediction. 9. The method of claim 8, wherein the method of claim 8 – The steps of uncertainty prediction include: 7 26 1272737 Voltage measurement application · 'The correction step, the temperature measurement value, the current measurement value and the 5 number in the model, the uncertainty is executed The method described in claim 9 is as follows: / 5 10 15 20 5 10 to calculate a gain coefficient; using the gain coefficient, the voltage measurement is calculated by X hi., Li Wei The value and the internal state prediction, the slave internal state prediction; and the use of the gain wire and the material to determine the qualitative prediction. t π 杈正不石 U· The steps of the algorithm described in claim 10 include: The method, wherein the execution-known correction (four) state prediction and the correction uncertainty prediction, 豕..., the juice method is required to repeat the implementation in the next period of time. 12. If the patent application scope is described in item U, at least - It is selected from the group consisting of one card and one for the second time; the group consisting of the device • and , the seven and the extended Kalman according to the application of the 4th patent scope of the application of the ancient soil energy ~ # π, 77 return method , where the increase i a hysteresis φ , a voltage polarization range shoulder magnetic strength, a resistance, a charge 晷 t valley, a polarization voltage time when electricity = combination coefficient, a hysteresis mixing coefficient, - magnetic hysteresis (four) hoist number and one The method of grouping the efficiency constants, wherein the method of performing the calculation includes: 14 the step of uncertainty prediction according to item 2 of the Shenqing patent scope further comprises: measuring a temperature; and 27!272737, w. The value, the amount of current in the -number and y speak 丨A value and the voltage, day, and number of 杈 type, the uncertainty prediction is performed. The value of the value is applied 15 · If the scope of the application of the patent scope The step of predicting the uncertainty of the uncertainty includes: / 'where the execution 1 - 5 determines a current measurement value; determines a voltage measurement value; and sets the current measurement value and the voltage measurement value set to perform the internal increase Volume status prediction. The mathematical model 16_, as described in the fifteenth aspect of the patent application, the step of uncertainty prediction includes: , wherein the performing a measurement of the temperature; and the measurement of the temperature, the measurement of the current The value and the voltage are, and in a mathematical model, the uncertainty prediction is performed. The pressure value is applied as described in item 16 of the patent application scope. The first step includes: /, the correction step calculates a gain coefficient; the quantity state pre-use - the gain coefficient, the voltage measurement value, and the internal measurement, Calculating a corrected internal state prediction; and 20 utilizing the gain coefficient and the uncertainty predictive prediction. The method of calculating a correction is not provided. The method of claim 17, wherein the step of performing an exclusive method comprises: using the corrected internal state prediction and the corrected uncertainty prediction to obtain the algorithm The predictions required for implementation are repeated in the next period. 28 1272737. For a method, please refer to the method described in item 18, wherein the algorithm is selected from a group consisting of a Kalman wave device and an extended Kalman filter. 5' The method of claim i, wherein the method further converges to a respective physical value by a more or more incremental state, respectively, wherein the method of claim 2G, wherein Step 10 2; - Λ — - Battery model output of the incremental state of the ideal convergence value. This includes the method described in item 21 of the scope of interest, wherein the step is more quantitative. AH is based on the measurement of the incremental state of the current measurement. The method of claim 22, further comprising a 15 battery r: Kalman filter or an extended Kalman filter, , q output and the measurement vector adjust an incremental state evaluation. Include: 24.----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- The unit of prediction; 2 is the (4) incremental state prediction and the single, 'factory, force-/3⁄4 wv early prediction of the non-shirt prediction, which is the unit of the execution-algorithm of the unit of the internal incremental state prediction, Wherein the algorithm repeats the steps performed by the unit to form the internal incremental state pre-29 1272737, the second forecast, and the unit for correcting the internal incremental state prediction and the H-predictive prediction.俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾 俾5 Electric, and also: the device described in claim 24, wherein the unit for forming the internal state prediction of the Dianchi includes: a unit for measuring a current; a unit for measuring a voltage; The parameter evaluation, the current measurement value, and the electrical measurement value, the set 10, the unit in the learning model to perform the internal incremental state prediction. 26. The device of claim 25, wherein the unit for predicting the uncertainty of the two-stage incremental state prediction comprises: wherein: the current measurement value and the voltage measurement value are applied to - A unit in the mathematical model to perform this uncertainty prediction. The device of claim 26, wherein the correction is within the device. The unit for predicting the state of the 卩 state and the prediction of the uncertainty includes: a unit for calculating a gain coefficient; - calculating a unit for correcting the internal incremental state prediction by using the gain coefficient, the voltage measurement value, and the internal measurement And the unit of deterministic prediction using the gain coefficient and the uncertainty prediction. . The ten-one-one correction is not 28. The unit of the loading algorithm as described in claim 27 includes: - Bu Hai executes the 30 1272737 2 using the & positive internal incremental state system and the correction uncertainty pre-fetching The algorithm is used to predict the order required for the implementation of the next-period money, such as the device described in Zhongqing Patent Model No. 28, wherein the algorithm is linked to >, - selected from one by one Kalman A group of filters and devices. 30. The attack described in claim 29, wherein the incremental form includes: a small one selected from a state of charge, a degree of voltage polarization, a degree of resistance, A group consisting of a charge capacity, a polarization voltage time constant, a polarization voltage mixing coefficient, a hysteresis mixing coefficient, a hysteresis phenomenon ratio constant, and an efficiency constant. 31. The device of claim 25, wherein the unit for forming an internal incremental state prediction of the battery further comprises: a unit for measuring a temperature; and 15 a temperature measurement value, the The current measurement and the voltage measurement are applied to a mathematical model to perform the internal state prediction. 32. The apparatus of claim 31, wherein the unit for forming an uncertainty prediction of the internal incremental state prediction of the battery comprises: 20 measuring a temperature measurement value, the current measurement The value and the voltage measurement are applied to a mathematical model to perform the prediction of the uncertainty. 33. The apparatus of claim 32, wherein the unit for correcting the internal incremental state prediction and the uncertainty prediction comprises: a unit for calculating a gain coefficient; 31 1272737 - utilizing the gain coefficient, The voltage measurement value and the internal shape ', the leaf nose correction unit for the internal incremental state prediction; and 丨. Determining the coefficient and the uncertainty prediction to calculate - the correction is not 5 10 34. The apparatus of claim 33, wherein the unit for performing the method includes: ·, using the corrected internal incremental state Predicting, correcting internal state predictions and correcting uncertainty predictions, and knowing the units of predictions required by the algorithm for the next time period. And less than - from - - - Kalman filter and an extended Kalman filter 35. The device of claim 34, wherein the algorithm comprises a group 36. The device of claim 35, wherein the incremental state comprises at least one selected from the group consisting of a state of charge, a voltage polarization, a degree of hysteresis, a resistance, a charge capacity, a polarization voltage time constant , a group of polarization voltage mixing coefficients, a hysteresis mixing coefficient, a hysteresis phenomenon ratio constant, and an efficiency constant. 37. The apparatus of claim 25, wherein the unit for forming an uncertainty prediction of the in-line state of the battery is further comprising: a unit for measuring a temperature; and a temperature The measured value, the current measured value, and the voltage measured value are applied to a mathematical model to perform the unit of the uncertainty prediction. 38. The apparatus of claim 24, wherein the unit for determining an uncertainty prediction of an internal incremental state prediction of the battery comprises: 32 ^72737 a unit for measuring a current; and a unit for measuring a voltage And 5 the current measurement value and the voltage amount to perform the uncertainty prediction. 'The value set is used for a mathematical model 10 39. The unit for predicting the uncertainty of the internal incremental state prediction of the device battery as described in claim 38 of the patent application further includes: " a unit for measuring the temperature; and applying四 (4) The uncertainty measurement is performed in the temperature measurement value, the current measurement value, and the voltage measurement value 'mathematical type'. The apparatus of the above-mentioned 4th, or the patent scope of claim 39, wherein the correction of the internal 4 reading of the '4 off and the uncertainty system comprises: a unit for calculating a gain coefficient; 15 n a profit! The gain coefficient, the voltage measurement value, and the internal increment state are used to calculate a unit for correcting the internal state prediction and a component using the gain coefficient and the uncertainty prediction to calculate a unit for correcting the uncertainty prediction. 41. The unit of the loading algorithm as described in the fourth paragraph of the patent application includes: "where the execution of the method is performed, the internal state prediction is corrected, the internal state prediction is corrected, and the correction uncertainty is determined. The prediction is made by knowing the unit of prediction required for the algorithm to repeat the implementation in the next period. The device of claim 41, wherein the algorithm 2 is at least one selected from the group consisting of a Kalman filter and an extended Kalman filter. 33 I272737 43. The apparatus of claim 24, further comprising a single 7/ 〇 44 in which the one or more incremental states respectively converge to respective physical values. The apparatus of the item, wherein the means for ensuring that the five or more incremental states respectively converge to respective corresponding physical values comprises a unit providing a battery model output having an incremental state of an ideal convergence value. 45. The apparatus of claim 44, wherein the means for ensuring that the - or multi-incremental state respectively converge to respective corresponding physical values further comprises providing a state based on the current one or more increments The unit of the corresponding measurement vector of the measured value. 46. The apparatus of claim 45, further comprising a Kalman filter or an extended Kalman filter' utilizing the battery model output and the measurement vector to adjust an incremental state assessment. 15 47. An apparatus for evaluating a current incremental state of an electrochemical cell: a device for forming an internal incremental state prediction of the pair of electrochemical cells, wherein the incremental state comprises at least one internal state value and at least An internal 20 value; the mouth sighs a pair of the internal incremental state prediction to form an uncertainty prediction, the internal incremental state prediction and the uncertainty preset; and the exhibition 1272737 one execution one An apparatus for algorithm, wherein the algorithm repeats the prediction of the internal incremental state, the prediction of the uncertainty of the internal incremental state prediction, and the correction of the internal incremental state prediction and the uncertainty prediction , 产生 generates an current estimate of the singular state and produces an existing uncertainty for the incremental state 5 evaluation. 48. A storage medium that uses a machine readable computer program code, Uma, a storage medium, and a computer that drives a computer to perform a method of evaluating the current incremental state of an electrochemical cell. The method includes: forming an internal incremental dispersion prediction for the electrochemical battery system, wherein the incremental state includes at least an internal state value and at least one internal parameter value; forming an uncertainty prediction for the internal incremental state prediction; Performing an algorithm, a method in which the a-history algorithm repeats the internal increment state 49· 一種藉由電腦資料訊號傳送之傳送媒體 校正該内部增量狀態預測及該不確定性預測;以及49. A transmission medium transmitted by a computer data signal corrects the internal incremental state prediction and the uncertainty prediction; 對該内部增量狀態預測形成-不確定性預測； 权正A内邛增量狀悲預測及該不確定性預測； 35 1272737 執行-演算法，其中該演算法係重複 預測、該不確定性預钏另兮於不兮咖A # &里狀心、 _及純正該㈣增量狀態預測及該 不確定性制，俾對該增量狀態產生—現行評估並對該增 量狀態評估產生一現行不確定性。 曰 36 1272737 七 指定代表圖： (一） 本案指定代表圖為：圖（1 )。 (二） 本代表圖之元件符號簡單說明： 20電池組 30負載電路 42電流感測器 46温度感測器 10參數評估系統 22電池 40量測裝置 44電壓感測器 48壓力感測器及/或阻抗感測器 50運算電路 52儲存媒體 54傳播訊號 八、本案若有化學式時，請揭示最能顯示發明特徵的化學式:The internal incremental state prediction formation-uncertainty prediction; the weighted positive A 邛 incremental sorrow prediction and the uncertainty prediction; 35 1272737 Execution-algorithm, wherein the algorithm is repeated prediction, the uncertainty The pre-existing 钏 钏 A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A current uncertainty.曰 36 1272737 VII Designated representative map: (1) The representative representative of the case is: Figure (1). (B) The symbol of the representative figure is briefly described: 20 battery pack 30 load circuit 42 current sensor 46 temperature sensor 10 parameter evaluation system 22 battery 40 measuring device 44 voltage sensor 48 pressure sensor and / Or the impedance sensor 50 operation circuit 52 stores the medium 54 to propagate the signal. 8. If there is a chemical formula in this case, please disclose the chemical formula that best shows the characteristics of the invention:
TW093136725A 2004-11-23 2004-11-29 Method and system for joint battery state and parameter estimation TWI272737B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US10/995,599 US7593821B2 (en) 2004-11-23 2004-11-23 Method and system for joint battery state and parameter estimation
PCT/KR2004/003102 WO2006057469A1 (en) 2004-11-29 2004-11-29 Method and system for joint battery stateand parameter estimation

Publications (1)

Publication Number Publication Date
TWI272737B true TWI272737B (en) 2007-02-01

Family

ID=36461976

Family Applications (1)

Application Number Title Priority Date Filing Date
TW093136725A TWI272737B (en) 2004-11-23 2004-11-29 Method and system for joint battery state and parameter estimation

Country Status (2)

Country Link
US (1) US7593821B2 (en)
TW (1) TWI272737B (en)

Families Citing this family (71)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7321220B2 (en) * 2003-11-20 2008-01-22 Lg Chem, Ltd. Method for calculating power capability of battery packs using advanced cell model predictive techniques
US9851414B2 (en) 2004-12-21 2017-12-26 Battelle Energy Alliance, Llc Energy storage cell impedance measuring apparatus, methods and related systems
US7723957B2 (en) * 2005-11-30 2010-05-25 Lg Chem, Ltd. System, method, and article of manufacture for determining an estimated battery parameter vector
FR2908322B1 (en) * 2006-11-09 2009-03-06 Parrot Sa METHOD FOR DEFINING GAMING AREA FOR VIDEO GAMING SYSTEM
FR2908324B1 (en) * 2006-11-09 2009-01-16 Parrot Sa DISPLAY ADJUSTMENT METHOD FOR VIDEO GAMING SYSTEM
FR2908323B1 (en) * 2006-11-09 2009-01-16 Parrot Sa METHOD OF DEFINING A COMMON REFERENTIAL FOR A VIDEO GAMING SYSTEM
JP4703593B2 (en) * 2007-03-23 2011-06-15 株式会社豊田中央研究所 Secondary battery state estimation device
US10379168B2 (en) 2007-07-05 2019-08-13 Battelle Energy Alliance, Llc Apparatuses and methods for testing electrochemical cells by measuring frequency response
US8628872B2 (en) * 2008-01-18 2014-01-14 Lg Chem, Ltd. Battery cell assembly and method for assembling the battery cell assembly
US7994755B2 (en) 2008-01-30 2011-08-09 Lg Chem, Ltd. System, method, and article of manufacture for determining an estimated battery cell module state
US9140501B2 (en) * 2008-06-30 2015-09-22 Lg Chem, Ltd. Battery module having a rubber cooling manifold
US8067111B2 (en) * 2008-06-30 2011-11-29 Lg Chem, Ltd. Battery module having battery cell assembly with heat exchanger
US8426050B2 (en) * 2008-06-30 2013-04-23 Lg Chem, Ltd. Battery module having cooling manifold and method for cooling battery module
US7883793B2 (en) * 2008-06-30 2011-02-08 Lg Chem, Ltd. Battery module having battery cell assemblies with alignment-coupling features
US9759495B2 (en) 2008-06-30 2017-09-12 Lg Chem, Ltd. Battery cell assembly having heat exchanger with serpentine flow path
US8202645B2 (en) 2008-10-06 2012-06-19 Lg Chem, Ltd. Battery cell assembly and method for assembling the battery cell assembly
US8116998B2 (en) 2009-01-30 2012-02-14 Bae Systems Controls, Inc. Battery health assessment estimator
US9030169B2 (en) * 2009-03-03 2015-05-12 Robert Bosch Gmbh Battery system and method for system state of charge determination
US9337456B2 (en) * 2009-04-20 2016-05-10 Lg Chem, Ltd. Frame member, frame assembly and battery cell assembly made therefrom and methods of making the same
US8852778B2 (en) * 2009-04-30 2014-10-07 Lg Chem, Ltd. Battery systems, battery modules, and method for cooling a battery module
US8663829B2 (en) 2009-04-30 2014-03-04 Lg Chem, Ltd. Battery systems, battery modules, and method for cooling a battery module
US8663828B2 (en) * 2009-04-30 2014-03-04 Lg Chem, Ltd. Battery systems, battery module, and method for cooling the battery module
US8403030B2 (en) 2009-04-30 2013-03-26 Lg Chem, Ltd. Cooling manifold
US8703318B2 (en) * 2009-07-29 2014-04-22 Lg Chem, Ltd. Battery module and method for cooling the battery module
US8399118B2 (en) * 2009-07-29 2013-03-19 Lg Chem, Ltd. Battery module and method for cooling the battery module
US8399119B2 (en) * 2009-08-28 2013-03-19 Lg Chem, Ltd. Battery module and method for cooling the battery module
US10204706B2 (en) * 2009-10-29 2019-02-12 Medtronic, Inc. User interface for optimizing energy management in a neurostimulation system
US8427105B2 (en) * 2009-12-02 2013-04-23 Gregory L. Plett System and method for equalizing a battery pack during a battery pack charging process
US8041522B2 (en) * 2009-12-02 2011-10-18 American Electric Vehicles, Ind. System and method for recursively estimating battery cell total capacity
US8918299B2 (en) * 2009-12-02 2014-12-23 American Electric Vehicles, Inc. System and method for maximizing a battery pack total energy metric
US20130069660A1 (en) * 2010-02-17 2013-03-21 Julien Bernard Method for in situ battery diagnostic by electrochemical impedance spectroscopy
US8341449B2 (en) 2010-04-16 2012-12-25 Lg Chem, Ltd. Battery management system and method for transferring data within the battery management system
US8942935B2 (en) * 2010-06-14 2015-01-27 Medtronic, Inc. Charge level measurement
FR2965360B1 (en) * 2010-09-27 2013-03-29 IFP Energies Nouvelles METHOD FOR IN SITU DIAGNOSIS OF BATTERIES BY SPECTROSCOPY OF ELECTROCHEMICAL IMPEDANCE
US8449998B2 (en) 2011-04-25 2013-05-28 Lg Chem, Ltd. Battery system and method for increasing an operational life of a battery cell
US10234512B2 (en) * 2011-06-11 2019-03-19 Sendyne Corporation Current-based cell modeling
US8706333B2 (en) * 2011-06-28 2014-04-22 Ford Global Technologies, Llc Nonlinear observer for battery state of charge estimation
US8859119B2 (en) 2011-06-30 2014-10-14 Lg Chem, Ltd. Heating system for a battery module and method of heating the battery module
US8993136B2 (en) 2011-06-30 2015-03-31 Lg Chem, Ltd. Heating system for a battery module and method of heating the battery module
US8974929B2 (en) 2011-06-30 2015-03-10 Lg Chem, Ltd. Heating system for a battery module and method of heating the battery module
US8974928B2 (en) 2011-06-30 2015-03-10 Lg Chem, Ltd. Heating system for a battery module and method of heating the battery module
KR101863036B1 (en) * 2011-11-30 2018-06-01 주식회사 실리콘웍스 Method for estimating the state of charge of battery and battery management system
US8922217B2 (en) * 2012-05-08 2014-12-30 GM Global Technology Operations LLC Battery state-of-charge observer
JP5393837B2 (en) * 2012-05-11 2014-01-22 カルソニックカンセイ株式会社 Battery charge rate estimation device
US9417270B2 (en) 2013-01-08 2016-08-16 GM Global Technology Operations LLC Systems and methods to capture and utilize temperature information in a battery system
US8935043B2 (en) 2013-01-29 2015-01-13 Ford Global Technologies, Llc Temperature compensated battery parameter estimation
US10664562B2 (en) * 2013-02-24 2020-05-26 Fairchild Semiconductor Corporation and University of Connecticut Battery state of charge tracking, equivalent circuit selection and benchmarking
US9575128B2 (en) * 2013-03-12 2017-02-21 GM Global Technology Operations LLC Battery state-of-charge estimation for hybrid and electric vehicles using extended kalman filter techniques
KR101632351B1 (en) 2013-10-14 2016-06-21 주식회사 엘지화학 Apparatus for estimating state of hybrid secondary battery and Method thereof
CN103558556B (en) * 2013-10-31 2016-02-03 重庆长安汽车股份有限公司 A kind of power battery SOH estimation method
JP6299187B2 (en) * 2013-11-29 2018-03-28 富士通株式会社 Estimation program, estimation method, and estimation apparatus
US9660299B2 (en) * 2013-12-10 2017-05-23 Southwest Research Institute Strain measurement based battery testing
US9197078B2 (en) * 2013-12-18 2015-11-24 Ford Global Technologies, Llc Battery parameter estimation
JP6455914B2 (en) * 2014-05-27 2019-01-23 学校法人立命館 Storage power remaining amount estimation device, method for estimating remaining power storage amount of storage battery, and computer program
EP2963434B1 (en) * 2014-06-30 2021-08-11 Foundation Of Soongsil University-Industry Cooperation Battery state estimation method and system using dual extended kalman filter, and recording medium for performing the method
US20160018468A1 (en) * 2014-07-21 2016-01-21 Richtek Technology Corporation Method of estimating the state of charge of a battery and system thereof
US9318778B2 (en) * 2014-09-17 2016-04-19 GM Global Technology Operations LLC Systems and methods for battery system temperature estimation
KR101767635B1 (en) 2014-10-24 2017-08-14 주식회사 엘지화학 Apparatus for estimating state of charge for secondary battery and Method thereof
FR3029298B1 (en) * 2014-11-28 2016-12-30 Renault Sa AUTOMATIC METHOD OF ESTIMATING THE CHARGING STATE OF A CELL OF A BATTERY
CN105068008B (en) * 2015-07-14 2018-10-19 南京航空航天大学 The battery charge state method of estimation of battery parameter is recognized using Vehicular charger
US10120035B2 (en) 2015-12-01 2018-11-06 Southwest Research Institute Monitoring and control of electrochemical cell degradation via strain based battery testing
CN105548893A (en) * 2015-12-07 2016-05-04 上海空间电源研究所 Method for describing and evaluating lithium ion battery health state
US10345384B2 (en) 2016-03-03 2019-07-09 Battelle Energy Alliance, Llc Device, system, and method for measuring internal impedance of a test battery using frequency response
US10656233B2 (en) 2016-04-25 2020-05-19 Dynexus Technology, Inc. Method of calibrating impedance measurements of a battery
CN105891729B (en) * 2016-06-23 2019-08-13 矽力杰半导体技术(杭州)有限公司 The condition detection method and device of battery and battery pack
CN108808137B (en) * 2018-06-19 2019-10-25 杭州电子科技大学 A kind of lithium battery management system
KR102465294B1 (en) 2019-01-23 2022-11-08 주식회사 엘지에너지솔루션 Battery management appratus, battery management method and battery pack
US11054481B2 (en) 2019-03-19 2021-07-06 Battelle Energy Alliance, Llc Multispectral impedance determination under dynamic load conditions
US12000902B2 (en) 2019-05-02 2024-06-04 Dynexus Technology, Inc. Multispectral impedance determination under dynamic load conditions
US11422102B2 (en) 2020-01-10 2022-08-23 Dynexus Technology, Inc. Multispectral impedance measurements across strings of interconnected cells
US11519969B2 (en) 2020-01-29 2022-12-06 Dynexus Technology, Inc. Cross spectral impedance assessment for cell qualification

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE10106505A1 (en) * 2001-02-13 2002-08-29 Bosch Gmbh Robert Method and device for condition detection of technical systems such as energy storage
US6534954B1 (en) * 2002-01-10 2003-03-18 Compact Power Inc. Method and apparatus for a battery state of charge estimator
US7109685B2 (en) * 2003-09-17 2006-09-19 General Motors Corporation Method for estimating states and parameters of an electrochemical cell
US8103485B2 (en) * 2004-11-11 2012-01-24 Lg Chem, Ltd. State and parameter estimation for an electrochemical cell
US7315789B2 (en) * 2004-11-23 2008-01-01 Lg Chem, Ltd. Method and system for battery parameter estimation

Also Published As

Publication number Publication date
US20060111870A1 (en) 2006-05-25
US7593821B2 (en) 2009-09-22

Similar Documents

Publication Publication Date Title
TWI272737B (en) Method and system for joint battery state and parameter estimation
Tran et al. A comprehensive equivalent circuit model for lithium-ion batteries, incorporating the effects of state of health, state of charge, and temperature on model parameters
TWI260808B (en) Apparatus and method for estimating stage of charge of battery using neural network
US7612532B2 (en) Method for controlling and monitoring using a state estimator having variable forgetting factors
US7197487B2 (en) Apparatus and method for estimating battery state of charge
US7315789B2 (en) Method and system for battery parameter estimation
US8185332B2 (en) Apparatus and method for estimating resistance characteristics of battery based on open circuit voltage estimated by battery voltage variation pattern
Farmann et al. Application-specific electrical characterization of high power batteries with lithium titanate anodes for electric vehicles
CN103502829B (en) For the optimization method of electrochemical storage system heat management
Yang et al. Battery states online estimation based on exponential decay particle swarm optimization and proportional-integral observer with a hybrid battery model
Zhang et al. A multi time-scale framework for state-of-charge and capacity estimation of lithium-ion battery under optimal operating temperature range
US20060100833A1 (en) State and parameter estimation for an electrochemical cell
Zhu et al. Accurate lithium-ion battery modeling with inverse repeat binary sequence for electric vehicle applications
EP4212896A1 (en) Method for estimating state of charge of battery
TWI287313B (en) Method and system for battery state and parameter estimation
US20220268856A1 (en) Method for detecting internal short-circuited cell
KR100878123B1 (en) Method and system for battery state and parameter estimation
CN102959791B (en) For determining the method for at least one state of multiple secondary battery unit, accumulator and motor vehicles
EP3988952B1 (en) Method for detecting abnormal battery cell
Selvabharathi et al. Experimental analysis on battery based health monitoring system for electric vehicle
JP5259190B2 (en) Joint battery condition and parameter estimation system and method
Luan et al. Research on variable time-scale SOC and SOH asynchronous collaborative estimation strategy for electric vehicle power lithium iron phosphate batteries
US20230176131A1 (en) Method for simulation of battery pack
Xu et al. The electric-thermal coupling simulation and state estimation of lithium-ion battery
KR20150034593A (en) Method and apparatus for state of charge estimation of battery